Code Coverage is Crucial in the IP Qualification Process
by Tom Borgstrom and Jimmy Awan
The Challenge of Verifying Configurable IP
Silicon intellectual property (IP) providers face a rising number of challenges in providing fully verified cores to their customers. IP users expect cores to be fully verified and robust in every usage scenario. In verifying their IP, providers must therefore try to anticipate each and every situation a core will encounter when integrated into a customer's system. Increased complexity and size mean a staggering number of test cases must be run to successfully verify an IP, and it must be done as quickly and economically as possible. This challenge is multiplied when the core in question is customer-configurable, with thousands of potential configurations.
Since configurable IP is so difficult to verify, why create it in the first place? The benefits are numerous. For the IP vendor, it means having just one master core to develop and maintain, instead of thousands of variant cores. For the IP user, there is the flexibility to reconfigure the core on the fly to meet the changing needs of project. If a design specification changes, requiring a change to the core, a traditional reusable core would have to be replaced, redesigned or designed from scratch. A configurable core, on the other hand, is easily adaptable. User-configurability also minimizes or eliminates the need to perform detailed RTL modifications by hand to meet design requirements. And once configured, these cores exactly meet SoC requirements.
When an IP vendor is developing configurable IP, they have to bear in mind not only the complexity of the task, but also the constraints of time and budget. Customers face shrinking market windows and the challenge of integrating their own IP with third party IP of which they have no first-hand knowledge. They need to know that a chip is fully verified and will be bug-free in all possible scenarios.
With these challenges, how can a vendor verify a core enough to be able to say with any degree of certainty that it will perform as expected in any configuration? Given unlimited time and resources, the vendor could create tests for each scenario. But we all know that time is anything but unlimited. And the vendor must not only make sure that the core is bug-free, but must also enable the customer to integrate the core into their system quickly and at the lowest cost possible.
The Challenge of Verifying Configurable IP
Silicon intellectual property (IP) providers face a rising number of challenges in providing fully verified cores to their customers. IP users expect cores to be fully verified and robust in every usage scenario. In verifying their IP, providers must therefore try to anticipate each and every situation a core will encounter when integrated into a customer's system. Increased complexity and size mean a staggering number of test cases must be run to successfully verify an IP, and it must be done as quickly and economically as possible. This challenge is multiplied when the core in question is customer-configurable, with thousands of potential configurations.
Since configurable IP is so difficult to verify, why create it in the first place? The benefits are numerous. For the IP vendor, it means having just one master core to develop and maintain, instead of thousands of variant cores. For the IP user, there is the flexibility to reconfigure the core on the fly to meet the changing needs of project. If a design specification changes, requiring a change to the core, a traditional reusable core would have to be replaced, redesigned or designed from scratch. A configurable core, on the other hand, is easily adaptable. User-configurability also minimizes or eliminates the need to perform detailed RTL modifications by hand to meet design requirements. And once configured, these cores exactly meet SoC requirements.
When an IP vendor is developing configurable IP, they have to bear in mind not only the complexity of the task, but also the constraints of time and budget. Customers face shrinking market windows and the challenge of integrating their own IP with third party IP of which they have no first-hand knowledge. They need to know that a chip is fully verified and will be bug-free in all possible scenarios.
With these challenges, how can a vendor verify a core enough to be able to say with any degree of certainty that it will perform as expected in any configuration? Given unlimited time and resources, the vendor could create tests for each scenario. But we all know that time is anything but unlimited. And the vendor must not only make sure that the core is bug-free, but must also enable the customer to integrate the core into their system quickly and at the lowest cost possible.
Related Semiconductor IP
- UCIe D2D Adapter & PHY Integrated IP
- Low Dropout (LDO) Regulator
- 16-Bit xSPI PSRAM PHY
- MIPI CSI-2 CSE2 Security Module
- ASIL B Compliant MIPI CSI-2 CSE2 Security Module
Related Articles
- CPF Based Verification of an SoC - Lessons Learnt
- Low Power Analysis and Verification of Super Speed Inter-Chip (SSIC) IP
- An Effective way to drastically reduce bug fixing time in SoC Verification
- Reduce SoC verification time through reuse in pre-silicon validation
Latest Articles
- RISC-V Functional Safety for Autonomous Automotive Systems: An Analytical Framework and Research Roadmap for ML-Assisted Certification
- Emulation-based System-on-Chip Security Verification: Challenges and Opportunities
- A 129FPS Full HD Real-Time Accelerator for 3D Gaussian Splatting
- SkipOPU: An FPGA-based Overlay Processor for Large Language Models with Dynamically Allocated Computation
- TensorPool: A 3D-Stacked 8.4TFLOPS/4.3W Many-Core Domain-Specific Processor for AI-Native Radio Access Networks